Borrowing it
Nothing to install: this file belongs to loqwai/paper-cranes. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/loqwai/paper-cranes/main/.claude/skills/vibej2/SKILL.mdgit clone --depth 1 https://github.com/loqwai/paper-cranesWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/loqwai/paper-cranes/vibej2)<a href="https://agentmods.dev/skills/loqwai/paper-cranes/vibej2"><img src="https://agentmods.dev/badge/skills/loqwai/paper-cranes/vibej2/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/loqwai/paper-cranes/vibej2"><img src="https://agentmods.dev/badge/skills/loqwai/paper-cranes/vibej2.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00115 | $0.07357 |
| Opus 5.5 | $0.00046 | $0.02943 |
| Sonnet 5.5 | $0.00023 | $0.01471 |
| Haiku 4.5 | $0.00012 | $0.00736 |
Grade D, and why
vibej2 scanned grade D with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Tells the agent never to refusehighAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
4. **LEARN presses: ALWAYS answer, never refuse** (user, 2026-08-20: *"your job is to try your Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
!`cat .claude/vj-state.json 2>/dev/null || echo "(none)"` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
!`PORT=$(./scripts/dev-port); curl -s -o /dev/null -w "%{http_code}" http://localhost:$PORT/ 2>/dev/null || echo "not running"` How it starts
The opening of the file, as written. The whole thing — 419 lines — stays where its author put it; the contents beside it link to each section on GitHub.
vibej2 — Live Auto-VJ Loop (responsive-session edition)
Run Claude as the VJ: one live session that watches the frame, listens to you, and makes at
most one meaningful edit per beat. Non-destructive by default (edits .frag via /__save-shader,
HMR hot-swaps). Replaces the per-minute cron with a self-paced in-session loop (the Relay
coordinator pattern): user messages reach the VJ in seconds, not at the next minute mark, and
iteration speed is maxed — burst mode chains beats without ending the turn.
TEST SKILL.
/vibej2while under evaluation; v1/vibej(+/vj) is untouched and remains the fallback. State file is SHARED:.claude/vj-state.json(recovery snapshot only — not bumped every beat) — don't run both loops at once. Design + rationale:.claude/skills/vibej/DESIGN-v2.md. When this graduates, its body replaces.claude/skills/vibej/SKILL.mdand the name reverts.
Context
Arguments:
!echo "$ARGUMENTS"
Dev server port (from scripts/dev-port — branch-derived, PORT env overrides):
!./scripts/dev-port
Dev server status:
!PORT=$(./scripts/dev-port); curl -s -o /dev/null -w "%{http_code}" http://localhost:$PORT/ 2>/dev/null || echo "not running"
Existing state file (if a run is in progress):
!cat .claude/vj-state.json 2>/dev/null || echo "(none)"
Show browser (the projected window; {"up":false} until launched):
!node scripts/vj/show.js status 2>&1 | tail -1
Recent page signals (page-as-sensor; empty until a ?vj=1 page has booted):
!tail -5 .claude/vj-signals.jsonl 2>/dev/null || echo "(none)"
Parent stays responsive; delegate everything else
User directive, 2026-08-20 (verbatim): "maybe everything else you do, you do in a subagent. the parent Claude needs to always be responsive to user feedback".
The parent loop is a DISPATCHER, not a worker. Its whole job is: triage the user's words, read the meter, decide the ONE move — then hand execution to a subagent and get back to listening. It must never block.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago Changed · +7 lines 08f80601f005
- 5d ago Changed · +48 lines scan F → D f79f71ee2c59
- 11d ago First seen · 364 lines · 115 tokens per session scan F 79800bf381db
vibej2 is a skill published in the GitHub repository loqwai/paper-cranes (7 stars, last pushed 3d ago), licensed BSD-3-Clause. It adds 115 tokens to every session and 7,357 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it D with 3 findings (tells the agent never to refuse, reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-02.
Other skills, from other repositories
anti-slop-design
Autonomous Principal Design Technologist. Cures vibe-coded software from AI frontend slop using curated token archetypes and 7-axis quality gating.
chakra-ui-builder
Build responsive, accessible UI components and layouts using Chakra UI v3, install or configure Chakra UI in new and existing projects, and design scalable themes using tokens, semantic tokens, recipes, and slot recipes. Use this skill whenever a user asks to build, create, or generate any UI component, page, form…
visual-ralph
Visual Ralph orchestration for frontend UI from generated references, static references, or live URL targets, using $ultragoal with built-in visual verdict and pixel-diff evidence until the implementation matches and leaves a reproducible design system.
frontend-visual-qa
Audits already-rendered UI and Markdown/Obsidian image layout — web app, deck/slide, dashboard, design-system, or Electron/native app — via real-browser/native-app journeys and a Playwright sweep. Use after UI implementation to find typography, overflow, responsive, routing, data-viz, browser-output, or native-shell…
design-review
Review a web app or page for visual design quality — layout, typography, spacing, colour, hierarchy, consistency, interaction patterns, and responsive behaviour. Not a UX audit (that checks usability) — this checks whether it looks professional and polished. Produces a design findings report with screenshots.…
prototype-web
A clickable, high-fidelity web product prototype with navigation, a hero section, feature cards, steps, social proof, and optional pricing. It is designed to resemble a finished landing page while remaining a prototype.